RESEARCH·BENCHMARK·2025
Benchmarking Humans & AI in Contract Drafting
A 13-tool benchmark of contract-drafting AI against a human-lawyer baseline — 450 outputs, 72 surveys, 12 interviews with in-house leaders.
/ Overview
“Is the AI good enough to draft this?” gets answered with vibes far more often than evidence. This study measures it: thirteen contract-drafting tools put up against a human-lawyer baseline, scored on the same tasks, with the people who actually live with the output weighing in.
It is less about declaring a winner than about giving in-house teams a defensible way to decide where AI drafting helps and where it quietly costs them later.
/ What it does
- Compares 13 contract-drafting tools against a human-lawyer baseline across a common task set.
- Collects 450 outputs, 72 structured surveys, and 12 interviews with in-house leaders.
- Separates where AI drafting saves real time from where it shifts cost into review.
/ Approach
Mixed-methods on purpose: the quantitative scores say what happened, the interviews say why it matters. Methodology and data are published so the conclusions can be checked rather than taken on trust — the same verification-first stance behind everything here.